Back to Lisa's Journal or Go to Journal 9
8 DVB
Linear regression is usually used to predict a dependent variable with an independent variable. Dee's example was using level of education to predict income. Multiple regression is where more than two variables are involved so there is an affect on the dependent variable by more than one thing. Again, Dee's example added in age as an additional predictor of income. Using regression you can answer questions that you could not with correlation. The relationship between the variables can be direct or indirect. the formula gives a starting point and then increments of change by the dependent variable with each increment of change by the independent variable. The slope of the line is set by these points on a graph. The strength of the predictor (independent variable) is expressed as R squared. So if the R2 value is 0.50 that means that 50% of the effect on the dependent variable change is caused by the independent variable change. or it "explains 50% of the variation of _variable".
8 DVB
Linear regression is usually used to predict a dependent variable with an independent variable. Dee's example was using level of education to predict income. Multiple regression is where more than two variables are involved so there is an affect on the dependent variable by more than one thing. Again, Dee's example added in age as an additional predictor of income. Using regression you can answer questions that you could not with correlation. The relationship between the variables can be direct or indirect. the formula gives a starting point and then increments of change by the dependent variable with each increment of change by the independent variable. The slope of the line is set by these points on a graph. The strength of the predictor (independent variable) is expressed as R squared. So if the R2 value is 0.50 that means that 50% of the effect on the dependent variable change is caused by the independent variable change. or it "explains 50% of the variation of _variable".